Fire-fighting unmanned aerial vehicle attitude control method, system, medium, product and equipment

Abstract

The invention belongs to the technical field of unmanned aerial vehicle attitude control. According to the fire-fighting unmanned aerial vehicle attitude control method and system, the medium, the product and the equipment, the process noise covariance and the measurement noise covariance of a self-adaptive Kalman filter are dynamically adjusted in combination with weight factors, and the self-adaptive Kalman filter determines the real state of the fire-fighting unmanned aerial vehicle based on sensing data and control input. A linear quadratic regulator is adopted to determine an optimal control gain according to the real state of the fire-fighting unmanned aerial vehicle, and based on an adaptive gain and the optimal control gain, final control input is determined to control a motor of the fire-fighting unmanned aerial vehicle to work; according to the method, the control robustness in a complex environment is remarkably improved, high-precision tracking and quick response are realized, the dependence on a precise model is reduced, the sensor interference resistance is enhanced, the expansibility is high, and the method is suitable for multiple scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of UAV attitude control, and particularly to a method, system, medium, product and device for controlling the attitude of a fire-fighting UAV. Background Art

[0002] The statements in this part only provide the background art related to the present invention, and do not necessarily constitute the prior art.

[0003] Due to its mobility, accuracy and ability of rapid response, UAVs are widely used in various industries. In the field of fire fighting and rescue, the application of UAV systems also has important significance. Especially today when fires occur frequently, high-rise fire-fighting UAVs can provide great assistance to the fire-fighting industry. However, the severe environment of the fire scene faced by fire-fighting UAVs will have a certain impact on the sensors of the UAVs. High temperature and smoke will cause abnormal conditions such as an increase in the sensor noise of the UAVs.

[0004] The LQG (Linear Quadratic Gaussian) controller adopted by current UAVs consists of a Kalman filter and an LQR controller (Linear Quadratic Regulator). The Kalman part is used to observe the optimal state, and the LQR controller part is used for state feedback. However, the traditional Kalman filter uses fixed parameters. If the noise statistical characteristics change, it cannot automatically adjust the parameters, resulting in estimation deviation; the LQR controller assumes that the system has no external disturbance and the model is accurate. However, in practice, external disturbances may lead to a decline in performance, and the LQR controller requires that all state variables can be directly measured, and an observer (such as a Kalman filter) is needed to estimate the unmeasurable states, and the observation error will reduce the control performance. Summary of the Invention

[0005] In order to solve the deficiencies of the prior art, the present invention provides a method, system, medium, product and device for controlling the attitude of a fire-fighting UAV, which significantly improves the control robustness in complex environments, realizes high-precision tracking and rapid response, reduces the dependence on an accurate model, enhances the anti-sensor interference ability, has strong scalability, and is applicable to multiple scenarios.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: In the first aspect, the present invention provides a method for controlling the attitude of a fire-fighting UAV.

[0007] A method for controlling the attitude of a fire-fighting UAV includes the following processes: Obtain the sensing data of the fire-fighting UAV; The process noise covariance and measurement noise covariance of an adaptive Kalman filter are dynamically adjusted in combination with weight factors, and the adaptive Kalman filter determines the true state of the fire-fighting UAV based on the sensing data and control input; A linear quadratic regulator is used to determine an optimal control gain according to the true state of the fire-fighting UAV, and based on the adaptive gain and the optimal control gain, a final control input is determined to control the operation of the motors of the fire-fighting UAV.

[0008] In a second aspect, the present invention provides a posture control system for a fire-fighting UAV.

[0009] A posture control system for a fire-fighting UAV includes: A data acquisition unit configured to: acquire sensing data of the fire-fighting UAV; A state generation unit configured to: dynamically adjust the process noise covariance and measurement noise covariance of an adaptive Kalman filter in combination with weight factors, and the adaptive Kalman filter determines the true state of the fire-fighting UAV based on the sensing data and control input; A posture control unit configured to: use a linear quadratic regulator to determine an optimal control gain according to the true state of the fire-fighting UAV, and based on the adaptive gain and the optimal control gain, determine a final control input to control the operation of the motors of the fire-fighting UAV.

[0010] In a third aspect, the present invention provides a computer device, including: a processor and a computer-readable storage medium; The processor is adapted to execute a computer program; The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, it implements the fire-fighting UAV posture control method as described in the first aspect of the present invention.

[0011] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, and the computer program is adapted to be loaded and executed by a processor to implement the fire-fighting UAV posture control method as described in the first aspect of the present invention.

[0012] In a fifth aspect, the present invention provides a computer program product, and the computer program product includes a computer program, and when the computer program is executed by a processor, it implements the fire-fighting UAV posture control method as described in the first aspect of the present invention.

[0013] Compared with the prior art, the beneficial effects of the present invention are: 1. The present invention significantly improves the control robustness in complex environments. In a fire field environment, the unmanned aerial vehicle (UAV) faces multiple interferences such as turbulence, thermal airflows, and smoke occlusion. Traditional controllers are prone to state estimation divergence or control instability due to fixed noise parameters and gains. The present invention can effectively suppress the influence of external disturbances on state estimation through the dynamic process and observation noise covariance matrix of the Adaptive Kalman Filter (AKF).

[0014] 2. The present invention realizes high-precision tracking and fast response. By adjusting the learning rate and increasing the learning rate in emergency environments to accelerate convergence, the system can make a faster and more accurate response.

[0015] 3. The present invention reduces the dependence on accurate models. Traditional LQR / LRG controllers rely on accurate dynamic models. The present invention enables the controller to stably track the attitude even when there are deviations in the model through the gain equation and AKF noise estimation.

[0016] 4. The present invention enhances the anti-sensor interference ability. The AKF dynamically corrects the observation noise. When the sensor is interfered, it can automatically adjust to reduce the sensor weight and maintain the stability of the system.

[0017] 5. The present invention has strong scalability and is applicable to multiple scenarios. It is not limited to quadrotor UAVs and fire field environments, but can also be extended to other dynamic systems and different interference environments.

[0018] Advantages of additional aspects of the present invention will be partly given in the following description, partly will become obvious from the following description, or will be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings forming a part of this specification are used to provide a further understanding of the present invention. The schematic embodiments and descriptions thereof of the present invention are used to explain the present invention and do not constitute an improper limitation of the present invention.

[0020] Figure 1 It is a schematic flow chart of the attitude control method of the fire-fighting UAV provided in Embodiment 1 of the present invention; Figure 2 It is a schematic principle framework diagram of the attitude control method of the fire-fighting UAV provided in Embodiment 1 of the present invention; Figure 3 It is a schematic diagram of a fire-fighting UAV attitude control system provided in Embodiment 2 of the present invention; Figure 4 It is a schematic diagram of a computer device provided in Embodiment 3 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0021] The present invention will be further described below in conjunction with the drawings and embodiments.

[0022] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0023] In the case of no conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other.

[0024] Embodiment 1: As described in the background art, traditional Kalman Filter (KF) and Linear Quadratic Gaussian (LQG) controllers assume that the process noise covariance and the measurement noise covariance are fixed values. However, in a dynamic environment such as a fire scene, the noise characteristics (such as turbulence and smoke occlusion) will change violently. In this implementation, through Adaptive Kalman Filter (AKF), the noise residuals are calculated in real time using a sliding window, and the process noise covariance matrix and the measurement noise covariance matrix are dynamically updated, which is more adaptable to sudden noise changes. And when the turbulence in the fire scene causes an increase or sensor failure causes a mutation, the state estimation accuracy can still be maintained, improving the real-time performance and robustness of the system.

[0025] Moreover, the traditional LQR controller is based on a fixed feedback gain , and it cannot cope with system parameter uncertainties. In this implementation, Model Reference Adaptive Control (MRAC) is introduced, and an adaptive gain is designed to dynamically adjust the control input. When facing disturbances in the fire scene, it is adjusted in real time to balance the response speed and stability. The integral term in the gain equation can suppress the steady-state error, and the differential term reduces the overshoot, improving the adaptive ability and disturbance resistance of the system.

[0026] This implementation integrates the gain adjustment of AKF (Adaptive Kalman Filter), LQR (Linear Quadratic Regulator), and adaptive control (MRAC) into a unified framework to form a closed-loop collaborative optimization. AKF provides high-precision state estimation and dynamically corrects noise parameters; LQR generates the optimal control input to minimize energy and tracking error; MRAC passes through Adaptive compensation for model uncertainty to enhance robustness. Each component is optimized independently, facilitating expansion. Meanwhile, the noise estimation of AKF provides a reliable state input for LQR, and MRAC further optimizes the control law, forming an "estimation - control - compensation" closed loop.

[0027] More specifically, this implementation proposes a method for attitude control of a fire - fighting drone, enabling the drone to adaptively adjust parameters when facing a fire scene to enhance the robustness and control performance of the system. Through dynamic noise adjustment, it can adapt to abnormal situations such as abnormal noise, as Figure 1 and Figure 2 shown, including the following processes: S1. Collect real - time data of the drone itself through the drone's own sensors (the drone is subject to external disturbances during operation), including information such as the drone's three - axis angular velocity, linear acceleration, position, speed, altitude, yaw angle, etc.; S2. The Adaptive Kalman Filter (AKF) is based on the drone sensor data. Through Kalman smoothing, it optimizes the historical state estimation and combines the control input to obtain the true state of the drone, and dynamically adjusts the process noise covariance matrix and the measurement noise covariance matrix ; S3. The LQR (Linear Quadratic Gaussian) controller obtains the optimal control gain by solving the Riccati equation and, based on the true state estimated by the AKF (Adaptive Kalman Filter), thus obtains the control input command; S4. By comparing with the output of the reference model (the input is the desired control command) to make an error, select an appropriate learning rate, and dynamically adjust the adaptive gain , and obtain a more accurate final control input; S5. The execution module of the fire - fighting drone (using a quad - rotor drone) converts the final control input into motor thrust to control the drone's attitude.

[0028] In S1 of this implementation, specifically, it includes: establishing a space inertial coordinate system and a body coordinate system , and the state vector of the fire - fighting drone can be expressed as: (1); where represents the position of the drone, represents the linear velocity of the drone on the , , three axes, and respectively represent the roll angle, pitch angle, and yaw angle of the drone, is the angular velocity on the three axes of the drone.

[0029] In S2 of this implementation method, the discretized linear model of the fire-fighting drone introducing disturbance factors such as fire field turbulence, hot air flow, and smoke is: (2); (3); Among them, is the state transition matrix, is the control input matrix, is the observation matrix, is the state value of the system at time, is the state value of the system at time, is the input value of the system at time, is the measured value of the system at time.

[0030] In equations (2) and (3), , are the total process noise and measurement noise respectively, and both follow a Gaussian distribution: (4); (5); Among them, represents the Gaussian distribution, represents the process noise generated by the system itself, represents the additional process noise introduced by external disturbances, represents the inherent observation noise of the system's built-in sensor itself, represents the observation noise caused by external disturbances. The Kalman filter approximates the influence of external disturbances through the covariance matrices , . Then the covariance matrix of the actual total noise is: (6); (7); Among them, is the total process noise covariance, is the basic process noise covariance (such as motor vibration, etc.), is the process noise covariance of the environmental increment, is the total observation noise covariance, is the basic observation noise covariance, representing the inherent error of the sensor, is the observation noise covariance of the environmental increment, represents is the covariance matrix of, denotes as the covariance matrix of.

[0031] For the process and measurement noise matrices, an adaptive dynamic adjustment method is adopted, and the steps are as follows: First, fixed-interval smoothing is used to obtain the smoothed state estimate , based on all the measurement data at the current time , to perform the optimal estimation of the state at the historical time , so as to use the future step observation data to correct the historical state estimate and reduce the influence of noise. Contrary to the Kalman filter (KF) estimating the current state , the fixed-interval smoothing estimate uses the measurement data at the current time and before to perform a retrospective estimation of the state at the past moments. Assuming the lag step size is , then at each time , state estimate values will be generated, , , …, corresponding to the current time and the past moments respectively.

[0032] Secondly, based on the smoothed state, calculate the process and measurement noise samples: (8); (9); where, is the state estimate at time, is the state estimate at time, is the control input at time, is the measurement value at time, After that, within the time window , statistically calculate the covariance of the process and measurement noise: (10); (11); where, (12); (13); where, represents the trace of a matrix, which is the sum of the elements on the main diagonal of a square matrix. is a coordinate transformation matrix for directionally weighting noise. is a diagonal matrix describing the degree of interference of the sensor. and are the process noise sample and the measurement noise sample respectively. , and are both covariance matrices, and they are covariance matrices after dynamic adjustment, that is, the covariance matrices updated in the previous iteration cycle: is a block matrix describing the state transition equation: (14); where represents the historical state transition matrix.

[0033] Subsequently, the state estimate is updated: (15); where is the Kalman gain calculated after update, is the observation matrix, is the state predicted for the current moment based on the state at the previous moment.

[0034] If encountering fire field turbulence, resulting in an increase in the process noise covariance , the predicted covariance expands, and the Kalman gain increases accordingly, and the AKF trusts the measurement value more; if encountering smoke blockage, resulting in an increase in a specific dimension of the observation noise covariance , the corresponding element of the Kalman gain decreases accordingly, and the AKF trusts the model prediction more, avoiding being misled by contaminated measurements.

[0035] Through the above design, the process and observation noise covariances and are dynamically adjusted to maintain the robust estimation of the adaptive Kalman filter AKF in the fire field environment and balance the model prediction and measurement trust.

[0036] In S3 and S4 of this implementation method, specifically, it includes: After the AKF estimates the state of the UAV, the LQR controller obtains the optimal control gain by solving the Riccati equation and obtains the control input command according to the true state estimated by the AKF.

[0037] Specifically, based on the state quantity estimated value obtained by the designed adaptive Kalman filter AKF, through the combination design of the LQR controller and the MRAC adaptive control rate, the robust LQG controller based on AKF is designed as follows: Set the optimal control index function of the UAV system model as follows: (16); where, is the state weight matrix, which is a symmetric positive semi - definite matrix; is the control weight matrix, which is a symmetric positive definite matrix; is the state of the system at time , and is the control input at time

[0038] Select a suitable weight matrix and solve for P through the Riccati equation: (17); where P is the matrix solution to be solved, is the state transition matrix, is the control input matrix.

[0039] Thus, the optimal control gain matrix is solved as: (18) ； The output of the LQR controller is then: (19); After that, by comparing with the output of the reference model to obtain the error, select a suitable learning rate, and dynamically adjust the gain ρ equation to obtain a more accurate final control input; Specifically, to improve the control effect of the LQR controller, combined with MRAC (Model Reference Adaptive Control), an adaptive gain ρ is introduced, and the control law becomes: (20); Introduce a second - order system reference model, and its transfer function is as follows: (21); where, is the natural frequency, is the damping ratio.

[0040] Make the measured data X output by the actual UAV track the ideal data output by the reference model, where X is the UAV state (such as the actual position, etc.) measured by the on - board sensors of the quad - rotor UAV, is the ideal state generated by the reference model according to the desired command, then the error can be obtained. To ensure that the error e is closer to 0, improve the gain ρ equation and let: (22); Wherein, is the learning rate, used to control the gain adjustment speed, and are the weight parameters for adjusting the error dynamics.

[0041] In S5 of this implementation manner, the execution module of the fire-fighting UAV converts the final control input into motor thrust to control the attitude of the UAV.

[0042] Embodiment 2: As Figure 3 shown, this implementation manner provides a fire-fighting UAV attitude control system, including: A data acquisition unit, configured to: acquire the sensing data of the fire-fighting UAV; A state generation unit, configured to: dynamically adjust the process noise covariance and measurement noise covariance of the adaptive Kalman filter in combination with weight factors, and the adaptive Kalman filter determines the true state of the fire-fighting UAV based on the sensing data and the control input; An attitude control unit, configured to: use a linear quadratic regulator to determine the optimal control gain according to the true state of the fire-fighting UAV, and determine the final control input based on the adaptive gain and the optimal control gain to control the operation of the motors of the fire-fighting UAV.

[0043] The specific working processes of the above units can be seen in the introduction in Embodiment 1, and will not be elaborated here.

[0044] It can be understood that the above units can be separately or all combined into one or several other units to form, or some of them can be further split into multiple smaller units in terms of function to form, which can achieve the same operation without affecting the realization of the technical effects of the embodiments of the present application. The above units are divided based on logical functions. In actual applications, the function of one unit can also be realized by multiple units, or the functions of multiple units can be realized by one unit. In other embodiments of the present application, the system may also include other units. In actual applications, these functions can also be assisted by other units and can be realized by the cooperation of multiple units.

[0045] According to another embodiment of the present application, the system described in this embodiment can be constructed and the method of Embodiment 1 of the present application can be implemented by running a computer program (including program code) capable of executing the respective steps involved in the corresponding method described in Embodiment 1 on a general computing device such as a computer including processing elements and storage elements such as a central processing unit (CPU), a random access memory (RAM), and a read-only memory (ROM). The computer program can be recorded on a computer-readable recording medium, for example, and loaded into the above computing device through the computer-readable recording medium and run therein.

[0046] Embodiment 3: As Figure 4 shown, this implementation provides an electronic device, which includes a processor 1001, a communication interface 1002, and a computer-readable storage medium 1003. Among them, the processor 1001, the communication interface 1002, and the computer-readable storage medium 1003 can be connected through a bus or other means.

[0047] Among them, the communication interface 1002 is used to receive and send data. The computer-readable storage medium 1003 can be stored in the memory of the electronic device. The computer-readable storage medium 1003 is used to store a computer program, and the computer program includes program instructions. The processor 1001 is used to execute the program instructions stored in the computer-readable storage medium 1003.

[0048] The processor 1001 (or CPU (Central Processing Unit, central processor)) is the computing core and control core of the electronic device, which is adapted to implement one or more instructions, and is specifically adapted to load and execute one or more instructions to implement the corresponding method flow or corresponding function.

[0049] The processor 1001 is configured to execute the following process: Obtain the sensing data of the fire-fighting drone; Dynamically adjust the process noise covariance and measurement noise covariance of the adaptive Kalman filter in combination with the weight factor. The adaptive Kalman filter determines the true state of the fire-fighting drone based on the sensing data and control input; Use a linear quadratic regulator to determine the optimal control gain according to the true state of the fire-fighting drone. Based on the adaptive gain and the optimal control gain, determine the final control input to control the operation of the motor of the fire-fighting drone.

[0050] For the specific working process, see the introduction in Embodiment 1 and will not be elaborated here.

[0051] Example 4: This implementation provides a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in an electronic device and is used to store programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the electronic device and, of course, the extended storage medium supported by the electronic device. The computer-readable storage medium provides a storage space, and this storage space stores the processing system of the electronic device.

[0052] Moreover, in this storage space, one or more instructions suitable for being loaded and executed by the processor are also stored. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory; optionally, it can also be at least one computer-readable storage medium located far from the aforementioned processor.

[0053] In one embodiment, one or more instructions are stored in the computer-readable storage medium; the processor loads and executes one or more instructions stored in the computer-readable storage medium to implement the following process: Obtain the sensing data of the fire-fighting drone; Dynamically adjust the process noise covariance and measurement noise covariance of the adaptive Kalman filter in combination with the weight factor. The adaptive Kalman filter determines the true state of the fire-fighting drone based on the sensing data and the control input; Use a linear quadratic regulator to determine the optimal control gain according to the true state of the fire-fighting drone. Based on the adaptive gain and the optimal control gain, determine the final control input to control the operation of the motors of the fire-fighting drone.

[0054] For the specific working process, see the introduction in Example 1 and will not be elaborated here.

[0055] Example 5: This implementation provides a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and these computer instructions are stored in a computer-readable storage medium. The processor of the electronic device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the electronic device to perform the following process: Obtain the sensing data of the fire-fighting drone; Dynamically adjust the process noise covariance and measurement noise covariance of the adaptive Kalman filter in combination with the weight factor. The adaptive Kalman filter determines the true state of the fire-fighting drone based on the sensing data and the control input; The linear quadratic regulator is used to determine the optimal control gain according to the actual state of the fire-fighting UAV, and based on the adaptive gain and the optimal control gain, the final control input is determined to control the operation of the motors of the fire-fighting UAV.

[0056] For the specific working process, please refer to the introduction in Embodiment 1 and will not be elaborated here.

[0057] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed in this application can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Skilled artisans can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0058] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions according to the embodiments of this application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted through a computer-readable storage medium. The computer instructions can be transmitted from a website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or a wireless manner (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data processing device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)).

[0059] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, various changes and modifications can be made to the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A firefighting drone attitude control method, characterized in that: The process includes: Obtain sensor data from firefighting drones; Dynamically adjusting the process noise covariance and the measurement noise covariance of an adaptive Kalman filter in combination with a weight factor, the adaptive Kalman filter determining the true state of the firefighting drone based on the sensor data and control input; A linear quadratic regulator is used to determine the optimal control gain according to the actual state of the fire-fighting drone, and based on the adaptive gain and the optimal control gain, a final control input is determined to control the motor operation of the fire-fighting drone.

2. The firefighting drone attitude control method according to claim 1, characterized in that: The sensor data of the firefighting drone includes: three-axis angular velocity, linear acceleration, position, speed, altitude and yaw angle of the firefighting drone.

3. The firefighting drone attitude control method according to claim 1, characterized in that: Dynamically adjust the process noise covariance of adaptive Kalman filter by combining weight factors and the measurement noise covariance ,include: ； ； in, , , is the time window, and are process noise samples and measurement noise samples respectively, represents the observation data for the next N steps, represent time, represent time, is the coordinate transformation matrix for directionally weighting the noise, is a diagonal matrix describing the degree of interference to the sensor, , is the block matrix describing the state transfer equation, is the observation matrix, is the measurement noise covariance, Represents the trace of the matrix.

4. The firefighting drone attitude control method according to claim 1, characterized in that: The actual status of the firefighting drone is: ； in, is the recalculated Kalman gain, For the system The measured value at the moment, For the system The observation matrix at time, For The real status of firefighting drones at the moment.

5. The firefighting drone attitude control method according to any one of claims 1 to 4, characterized in that: The output of the linear quadratic regulator for: ,in, is the adaptive gain, is the optimal control gain determined using a linear quadratic regulator, This is the actual status of the firefighting drone.

6. The firefighting drone attitude control method according to claim 5, characterized in that: Adaptive Gain Calculations include: ； in, is the learning rate, and To adjust the weight parameter of the error dynamics, is the reference model output, is the error, , Error The first derivative of Measurement data output for actual drones.

7. A firefighting drone attitude control system, characterized in that: include: The data acquisition unit is configured to: acquire sensor data of the firefighting drone; A state generation unit is configured to: dynamically adjust a process noise covariance and a measurement noise covariance of an adaptive Kalman filter in combination with a weight factor, wherein the adaptive Kalman filter determines a true state of the firefighting drone based on the sensor data and control input; The attitude control unit is configured to: use a linear quadratic regulator to determine the optimal control gain according to the actual state of the fire-fighting drone, and determine the final control input based on the adaptive gain and the optimal control gain to control the motor operation of the fire-fighting drone.

8. A computer device, characterized in that: include: A processor and a computer readable storage medium; a processor adapted to execute a computer program; A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by the processor, the firefighting drone attitude control method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is suitable for being loaded by a processor and executing the firefighting drone attitude control method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, it implements the fire-fighting drone attitude control method according to any one of claims 1 to 6.

Images